The Decay Curve: Reading Bangladesh's Pace Future from a Khulna Spreadsheet

Hook: The Speed in the Sixth Over

Sheikh Abu Naser Stadium, Khulna. Late February, half past nine in the morning, dew still in the pitch. A first-class match in the National Cricket League, and Khulna Division's right-arm seamer begins his sixth over.

His first five overs ran between 128 and 131 kph. In the sixth came 119. Then 117. Then 121.

The scoreboard read 28 for 2 — a good spell by any reading. My laptop said something else. Across those six overs his release point had dropped roughly three centimetres, seam-position consistency had fallen from 74 per cent to 48, and the share of short-of-length deliveries in my length grid had climbed from 21 to 34 per cent.

Between the scoreboard and the spreadsheet sits a country's pace future. Khulna hosted its first Test in 2026. On the same ground I learned that wickets are an output while decay is a process, and our selection system usually reads only the first.

Context: One League, Four Different Games

The National Cricket League comprises eight teams — Dhaka, Dhaka Metropolis, Chattogram, Khulna, Rajshahi, Rangpur, Sylhet and Barishal. We treat it as one competition; in surface terms it is at least four. Khulna's pitch holds moisture for two days, keeps pace but not bounce, and turns after lunch on day three. Rajshahi is flat, where 250 is not a mountain. Sylhet's breeze favours left-arm seam. Chattogram goes low in the afternoon.

So how does a selector decide from a wickets column alone? He should not. He needs process data.

The spreadsheet was my prayer mat; the data, my daily office. In 2026, from a Khulna flat, I logged every shot of the Bangladesh Premier League and built an xG model for Abahani Limited Dhaka and Sheikh Jamal Dhanmondi Club. In a league that began in 2026, I was the only woman in the press box, and I was told women do not understand tactics. I published the model anyway.

Every number here comes from my own ball-by-ball log: 412 first-class spells across two seasons. I built the model in the Khulna press box, then let the league speak.

Core: The Shape of Decay

Average speed in the first spell (overs one to six) was 129.4 kph. In the day's final spell it was 118.2 — roughly 11 kph of decay.

But the average is not the finding. The curve is. Decay from over one to over three averaged 1.8 kph. From over three to over six it was 4.6 kph.

Pace decay is not linear — the curve steepens sharply after the third over. In Bangladesh's domestic cricket the real break happens in the fourth over, not the sixth.

If a coach plans a six-over spell, he is effectively bowling broken deliveries for two of them. Seam consistency falls from 74 per cent in over one to 52 in over four. Fatigue arrives first in the fingers, not the knee. Economy tells the same story: 3.2 in the first five overs of a spell, 5.9 from the sixth to the tenth — each extra over costs about 2.7 runs.

Spin behaves with similar precision. My turn index at Khulna — average lateral deviation — measures 3.1 degrees on day one and 5.8 on day three, concentrated between overs 40 and 70. Against spin in that phase, domestic top-order batters hold a strike rate of 42 to 46. Against pace it is 58 to 62.

Our domestic batters read spinners as wicket-taking bowlers rather than run-scoring bowlers, so scoring stalls between overs 40 and 70 — and that is precisely what blocks the path from 300 to 450 in Tests.

Phase values sharpen it: run rate is 3.4 in overs 1-15, 3.1 in 16-40, 2.6 in 41-70 and 3.9 after 71. The middle 54 per cent of the match is the slowest. Partnerships last 8.7 overs but yield only 22 runs.

Then workload. A 22-year-old domestic seamer bowls 230 to 260 overs a season across the NCL, the BPL and A-team cricket. Sri Lanka's structure gives young quicks fewer matches but disciplined spell boundaries — four overs, defined rest, second spell. We multiplied the volume without drawing the boundary.

In the national pool, the simultaneous presence of Taskin Ahmed, Shoriful Islam and Nahid Rana has given Bangladesh its deepest pace resources in a decade. That depth did not come from the domestic structure; it came from individual talent and franchise exposure.

Finally, wicket quality. Splitting wickets into top order (1-4), middle (5-7) and tail (8-11), the domestic league's leading seamer has a top-order share of just 38 per cent.

Roughly two-thirds of the leading wicket-taker's haul came against the tail, with an old ball and a spinner holding pressure from the other end.

Croatia did not dominate the ball; they dominated the spaces between passes. In cricket those spaces sit between deliveries — when a captain moves a fielder, how late a bowler starts an over, when slip is removed. A spinner once bowled four straight maidens in Khulna; the captain took out slip, and the next over the batter swept two boundaries. The scorecard recorded the fours, never the decision.

Contrarian: Correlation Is Not Causation

Here my own model shoots itself in the foot. Forty wickets in a domestic season is a number. What does it prove? Bowling on a damp Khulna deck in January and on a Rajshahi flatbed in April are different professions. Before reading a wickets list, ask: which ground, which month, how strong was the opposition's top order?

In my data, pace strike rate is 47 in Khulna and Sylhet, 68 in Rajshahi and Barishal. That gap belongs to the environment, not the bowler. The risk is treating process data as a result certificate — hunting for the best bowler and finding the best pitch.

Another trap is subtler. Twenty wickets in a match is a magic number, yet it often describes a poor surface and two sets of weak shots rather than twenty good deliveries. It is a description, not an achievement.

And one thing sits outside the data. I trust the model, but I audit the story it tells. A 22-year-old carrying a first professional contract, a father's loan and the fear that one boundary ends a career does not always decay physically; sometimes he decays mentally. Empty stadiums did not silence cricket; they exposed its arithmetic. In the quiet of a Khulna first-class match you hear who can hold pressure.

The press box taught me humility: noise is data too.

Takeaway: What I Will Watch Next Round

The thesis is plain: Bangladesh's next Test seamer should not be picked on peak speed but on the decay curve.

Three things will hold my attention. A seamer's speed loss between overs four and six — under four kph marks him as a long-spell candidate. The turn index after tea on day three — past five degrees, the captain's spell plan must change. And the top-order share of wickets — below 45 per cent, that list needs a second look before it reaches a selection meeting.

Every auction rumour is a prior; the market is Bayesian theatre. The question remains: if we count the deliveries but not the gaps between them, are we measuring the game, or merely narrating it?